{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,23]],"date-time":"2026-06-23T03:58:54Z","timestamp":1782187134737,"version":"3.54.5"},"reference-count":30,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2020,7,29]],"date-time":"2020-07-29T00:00:00Z","timestamp":1595980800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2020,7,29]],"date-time":"2020-07-29T00:00:00Z","timestamp":1595980800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Imaging"],"published-print":{"date-parts":[[2020,12]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n<jats:title>Background<\/jats:title>\n<jats:p>This paper addresses issues of brain tumor, glioma, classification from four modalities of Magnetic Resonance Image (MRI) scans (i.e., T1 weighted MRI, T1 weighted MRI with contrast-enhanced, T2 weighted MRI and FLAIR). Currently, many available glioma datasets often contain some unlabeled brain scans, and many datasets are moderate in size.<\/jats:p>\n<\/jats:sec><jats:sec>\n<jats:title>Methods<\/jats:title>\n<jats:p>We propose to exploit deep semi-supervised learning to make full use of the unlabeled data. Deep CNN features were incorporated into a new graph-based semi-supervised learning framework for learning the labels of the unlabeled data, where a new 3D-2D consistent constraint is added to make consistent classifications for the 2D slices from the same 3D brain scan. A deep-learning classifier is then trained to classify different glioma types using both labeled and unlabeled data with estimated labels. To alleviate the overfitting caused by moderate-size datasets, synthetic MRIs generated by Generative Adversarial Networks (GANs) are added in the training of CNNs.<\/jats:p>\n<\/jats:sec><jats:sec>\n<jats:title>Results<\/jats:title>\n<jats:p>The proposed scheme has been tested on two glioma datasets, TCGA dataset for IDH-mutation prediction (molecular-based glioma subtype classification) and MICCAI dataset for glioma grading. Our results have shown good performance (with test accuracies 86.53% on TCGA dataset and 90.70% on MICCAI dataset).<\/jats:p>\n<\/jats:sec><jats:sec>\n<jats:title>Conclusions<\/jats:title>\n<jats:p>The proposed scheme is effective for glioma IDH-mutation prediction and glioma grading, and its performance is comparable to the state-of-the-art.<\/jats:p>\n<\/jats:sec>","DOI":"10.1186\/s12880-020-00485-0","type":"journal-article","created":{"date-parts":[[2020,7,29]],"date-time":"2020-07-29T12:02:39Z","timestamp":1596024159000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":85,"title":["Deep semi-supervised learning for brain tumor classification"],"prefix":"10.1186","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3521-8021","authenticated-orcid":false,"given":"Chenjie","family":"Ge","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Irene Yu-Hua","family":"Gu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Asgeir Store","family":"Jakola","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jie","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2020,7,29]]},"reference":[{"issue":"1","key":"485_CR1","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1186\/s12880-017-0198-4","volume":"17","author":"N Sauwen","year":"2017","unstructured":"Sauwen N, Acou M, Sima D, et al.Semi-automated brain tumor segmentation on multi-parametric mri using regularized non-negative matrix factorization. BMC Med Imaging. 2017; 17(1):29.","journal-title":"BMC Med Imaging"},{"issue":"1","key":"485_CR2","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1186\/s12880-019-0348-y","volume":"19","author":"R Sun","year":"2019","unstructured":"Sun R, Wang K, Guo L, et al.A potential field segmentation based method for tumor segmentation on multi-parametric mri of glioma cancer patients. BMC Med Imaging. 2019; 19(1):48.","journal-title":"BMC Med Imaging"},{"issue":"1","key":"485_CR3","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s12880-018-0301-5","volume":"19","author":"Y Wu","year":"2019","unstructured":"Wu Y, Zhao Z, Wu W, Lin Y, Wang M. Automatic glioma segmentation based on adaptive superpixel. BMC Med Imaging. 2019; 19(1):1\u201314.","journal-title":"BMC Med Imaging"},{"issue":"12","key":"485_CR4","doi-asserted-by":"crossref","first-page":"613","DOI":"10.1016\/j.cancergen.2012.10.009","volume":"205","author":"M Goodenberger","year":"2012","unstructured":"Goodenberger M, Jenkins R. Genetics of adult glioma. Cancer Genet. 2012; 205(12):613\u201321.","journal-title":"Cancer Genet"},{"issue":"5897","key":"485_CR5","doi-asserted-by":"crossref","first-page":"1807","DOI":"10.1126\/science.1164382","volume":"321","author":"D Parsons","year":"2008","unstructured":"Parsons D, Jones S, Zhang X, et al.An integrated genomic analysis of human glioblastoma multiforme. Science. 2008; 321(5897):1807\u2013812.","journal-title":"Science"},{"issue":"5","key":"485_CR6","doi-asserted-by":"crossref","first-page":"968","DOI":"10.3390\/molecules24050968","volume":"24","author":"B Kaminska","year":"2019","unstructured":"Kaminska B, Czapski B, Guzik R, et al.Consequences of idh1\/2 mutations in gliomas and an assessment of inhibitors targeting mutated idh proteins. Molecules. 2019; 24(5):968.","journal-title":"Molecules"},{"issue":"6","key":"485_CR7","doi-asserted-by":"crossref","first-page":"707","DOI":"10.1007\/s00401-010-0781-z","volume":"120","author":"C Hartmann","year":"2010","unstructured":"Hartmann C, Hentschel B, Wick W, et al.Patients with idh1 wild type anaplastic astrocytomas exhibit worse prognosis than idh1-mutated glioblastomas, and idh1 mutation status accounts for the unfavorable prognostic effect of higher age: implications for classification of gliomas. Acta neuropathol. 2010; 120(6):707\u201318.","journal-title":"Acta neuropathol"},{"issue":"8","key":"485_CR8","doi-asserted-by":"crossref","first-page":"1942","DOI":"10.1093\/annonc\/mdx230","volume":"28","author":"A Jakola","year":"2017","unstructured":"Jakola A, Skjulsvik A, Myrmel K, et al.Surgical resection versus watchful waiting in low-grade gliomas. Ann Oncol. 2017; 28(8):1942\u20138.","journal-title":"Ann Oncol"},{"issue":"8","key":"485_CR9","doi-asserted-by":"crossref","first-page":"765","DOI":"10.1056\/NEJMoa0808710","volume":"360","author":"H Yan","year":"2009","unstructured":"Yan H, Parsons D, Jin G, et al.Idh1 and idh2 mutations in gliomas. N Engl J Med. 2009; 360(8):765\u201373.","journal-title":"N Engl J Med"},{"issue":"3","key":"485_CR10","doi-asserted-by":"crossref","first-page":"882","DOI":"10.1148\/radiol.11110686","volume":"261","author":"Y Kang","year":"2011","unstructured":"Kang Y, Choi S, Kim Y, et al.Gliomas: histogram analysis of apparent diffusion coefficient maps with standard-or high-b-value diffusion-weighted mr imaging-correlation with tumor grade. Radiology. 2011; 261(3):882\u2013890.","journal-title":"Radiology"},{"issue":"7","key":"485_CR11","doi-asserted-by":"crossref","first-page":"1349","DOI":"10.3174\/ajnr.A2950","volume":"33","author":"J Carrillo","year":"2012","unstructured":"Carrillo J, Lai A, et al.Relationship between tumor enhancement, edema, idh1 mutational status, mgmt promoter methylation, and survival in glioblastoma. Am J Neuroradiol. 2012; 33(7):1349\u2013355.","journal-title":"Am J Neuroradiol"},{"issue":"6","key":"485_CR12","doi-asserted-by":"crossref","first-page":"1895","DOI":"10.3892\/ol.2014.2013","volume":"7","author":"S Qi","year":"2014","unstructured":"Qi S, Yu L, Li H, Ou Y, et al.Isocitrate dehydrogenase mutation is associated with tumor location and magnetic resonance imaging characteristics in astrocytic neoplasms. Oncol Lett. 2014; 7(6):1895\u2013902.","journal-title":"Oncol Lett"},{"issue":"8","key":"485_CR13","doi-asserted-by":"crossref","first-page":"3509","DOI":"10.1007\/s00330-016-4653-3","volume":"27","author":"J Yu","year":"2017","unstructured":"Yu J, Shi Z, Lian Y, Li Z, et al.Noninvasive idh1 mutation estimation based on a quantitative radiomics approach for grade ii glioma. Eur Radiol. 2017; 27(8):3509\u201322.","journal-title":"Eur Radiol"},{"issue":"4","key":"485_CR14","doi-asserted-by":"publisher","first-page":"916","DOI":"10.1002\/jmri.25960","volume":"48","author":"X Zhang","year":"2018","unstructured":"Zhang X, Tian Q, Wang L, Liu Y, et al.Radiomics strategy for molecular subtype stratification of lower-grade glioma: Detecting idh and tp53 mutations based on multimodal mri. J Magn Reson Imaging. 2018; 48(4):916\u201326. https:\/\/doi.org\/10.1002\/jmri.25960.","journal-title":"J Magn Reson Imaging"},{"issue":"4","key":"485_CR15","doi-asserted-by":"crossref","first-page":"563","DOI":"10.1007\/s11548-017-1691-5","volume":"13","author":"B Shofty","year":"2018","unstructured":"Shofty B, Artzi M, Bashat D, et al.Mri radiomics analysis of molecular alterations in low-grade gliomas. Int J CARS. 2018; 13(4):563\u201371.","journal-title":"Int J CARS"},{"issue":"2","key":"485_CR16","doi-asserted-by":"crossref","first-page":"299","DOI":"10.1007\/s11060-019-03096-0","volume":"142","author":"H Zhou","year":"2019","unstructured":"Zhou H, Chang K, Bai H, et al.Machine learning reveals multimodal mri patterns predictive of isocitrate dehydrogenase and 1p\/19q status in diffuse low-and high-grade gliomas. J Neuro-Oncol. 2019; 142(2):299\u2013307.","journal-title":"J Neuro-Oncol"},{"issue":"1","key":"485_CR17","doi-asserted-by":"crossref","first-page":"5467","DOI":"10.1038\/s41598-017-05848-2","volume":"7","author":"Z Li","year":"2017","unstructured":"Li Z, Wang Y, Yu J, et al.Deep learning based radiomics (dlr) and its usage in noninvasive idh1 prediction for low grade glioma. Sci Rep. 2017; 7(1):5467.","journal-title":"Sci Rep"},{"issue":"5","key":"485_CR18","doi-asserted-by":"crossref","first-page":"1073","DOI":"10.1158\/1078-0432.CCR-17-2236","volume":"24","author":"K Chang","year":"2018","unstructured":"Chang K, Bai H, Zhou H, Su C, et al.Residual convolutional neural network for the determination of idh status in low-and high-grade gliomas from mr imaging. Clin Cancer Res. 2018; 24(5):1073\u201381.","journal-title":"Clin Cancer Res"},{"issue":"8","key":"485_CR19","doi-asserted-by":"crossref","first-page":"382","DOI":"10.3390\/genes9080382","volume":"9","author":"S Liang","year":"2018","unstructured":"Liang S, Zhang R, Liang D, Song T, Ai T, Xia C, Xia L, Wang Y. Multimodal 3d densenet for idh genotype prediction in gliomas. Genes. 2018; 9(8):382.","journal-title":"Genes"},{"key":"485_CR20","doi-asserted-by":"crossref","first-page":"22560","DOI":"10.1109\/ACCESS.2020.2969805","volume":"8","author":"C Ge","year":"2020","unstructured":"Ge C, Gu I, Jakola A, Yang J. Enlarged training dataset by pairwise gans for molecular-based brain tumor classification. IEEE Access. 2020; 8:22560\u201370.","journal-title":"IEEE Access"},{"key":"485_CR21","unstructured":"Zhou D, Bousquet O, Lal T, Weston J, Sch\u00f6lkopf B. Learning with local and global consistency. In: Advances in Neural Information Processing Systems: 2004. p. 321\u2013328."},{"key":"485_CR22","doi-asserted-by":"crossref","unstructured":"Iscen A, Tolias G, Avrithis Y, Chum O. Label propagation for deep semi-supervised learning. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition: 2019. p. 5070\u20135079.","DOI":"10.1109\/CVPR.2019.00521"},{"key":"485_CR23","unstructured":"Keras. version 2.2.4, https:\/\/github.com\/fchollet\/keras. Accessed 03 Oct 2018."},{"key":"485_CR24","unstructured":"TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. Version 1.14.0, https:\/\/www.tensorflow.org\/. Accessed 19 June 2019."},{"key":"485_CR25","doi-asserted-by":"publisher","unstructured":"Segmentation Labels and Radiomic Features for the Pre-operative Scans of the TCGA-GBM Collection. 2017. https:\/\/doi.org\/10.7937\/K9\/TCIA.2017.KLXWJJ1Q.","DOI":"10.7937\/K9\/TCIA.2017.KLXWJJ1Q"},{"key":"485_CR26","doi-asserted-by":"publisher","unstructured":"Segmentation Labels and Radiomic Features for the Pre-operative Scans of the TCGA-LGG Collection. 2017. https:\/\/doi.org\/10.7937\/K9\/TCIA.2017.GJQ7R0EF.","DOI":"10.7937\/K9\/TCIA.2017.GJQ7R0EF"},{"issue":"10","key":"485_CR27","doi-asserted-by":"crossref","first-page":"1993","DOI":"10.1109\/TMI.2014.2377694","volume":"34","author":"B Menze","year":"2015","unstructured":"Menze B, Jakab A, et al.The multimodal brain tumor image segmentation benchmark (brats). IEEE Trans Med Imaging. 2015; 34(10):1993\u20132024.","journal-title":"IEEE Trans Med Imaging"},{"key":"485_CR28","doi-asserted-by":"crossref","first-page":"170117","DOI":"10.1038\/sdata.2017.117","volume":"4","author":"S Bakas","year":"2017","unstructured":"Bakas S, Akbari H, et al.Advancing the cancer genome atlas glioma mri collections with expert segmentation labels and radiomic features. Sci Data. 2017; 4:170117.","journal-title":"Sci Data"},{"key":"485_CR29","doi-asserted-by":"publisher","unstructured":"Pan Y, Huang W, et al.Brain tumor grading based on neural networks and convolutional neural networks. In: 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE: 2015. https:\/\/doi.org\/10.1109\/embc.2015.7318458.","DOI":"10.1109\/embc.2015.7318458"},{"key":"485_CR30","doi-asserted-by":"publisher","unstructured":"Ge C, Qu Q, Gu I, Jakola A. 3d multi-scale convolutional networks for glioma grading using mr images. In: 2018 25th IEEE International Conference on Image Processing (ICIP). IEEE: 2018. https:\/\/doi.org\/10.1109\/icip.2018.8451682.","DOI":"10.1109\/icip.2018.8451682"}],"container-title":["BMC Medical Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-020-00485-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12880-020-00485-0\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-020-00485-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,7,28]],"date-time":"2021-07-28T23:34:23Z","timestamp":1627515263000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcmedimaging.biomedcentral.com\/articles\/10.1186\/s12880-020-00485-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,7,29]]},"references-count":30,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2020,12]]}},"alternative-id":["485"],"URL":"https:\/\/doi.org\/10.1186\/s12880-020-00485-0","relation":{},"ISSN":["1471-2342"],"issn-type":[{"value":"1471-2342","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,7,29]]},"assertion":[{"value":"3 May 2020","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 July 2020","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 July 2020","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"This research has been approved by ethical committee of Western Sweden (Dnr: 702-18). Since data is from open data sources, any additional informed consent as part of this study was not appropriate, and the need for informed consent was waived by the ethics committee.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare that they have no competing interests.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"87"}}